amplitude but twice that value. 



This equation for the present should be considered a rough 

 approximation only, since the actual data were so scarce and 

 so infrequent that a true approximation of a continuous curve 

 was not very often obtained. In order to test the equation, a 

 few observations were available from weather ship NECTAR other 

 than those used in computing the constants. A listing of the 

 computed and observed heights (twice amplitude) follows: 

 Computed Observed 



28 26 



21 15 



33 36 



32 27 



105 1+8 



93 32 



The correlation coefficient between the above computed and observed 

 heights is 0.69, which, for only six cases, is of no statistical 

 significance. The results, however, seem to indicate a reasonable 

 basis for further work on this type of approach. The information 

 contained in equation (23) may be summarized by a single nomogram, 

 as shown in figure 26, For a given density of the upper 

 layer and for a given mean depth of the thermocline, the am- 

 plitude of the wave along the thermocline may be re^d off. 

 An unusual feature of this equation is the fact that there 

 exists a thermocline depth where the amplitude is independent 

 of density, and remains nearly constant. No particular physical 

 interpretation of this feature can be given. It should be 

 emphasized once again that equation (23) and figure 26 are 



55 



